Embed Size (px)
Citation preview
5
Musica Iagellonica 2013 ISSN 1233-9679
Marina ToffeTTI (Padova)
Restoring a Masterpiece. Some Remarks on the Reconstruction
of the Missing Part in Girolamo Frescobaldi’s Liber secundus diversarum modulationum (Rome, 1627)
This article sets out some reflections resulting from the reconstruction of the missing part of the cantus secundus in the Liber secundus diversarum modu lationum by Girolamo frescobaldi. 1 of the Liber secundus a single copy is preserved, now kept at the British Library in London and consisting of the part-books of the cantus primus, tenor, bassus and bassus ad organum. 2 The copy lacks one part-book (probably that of the cantus secundus) since the acquisi-tion by the British Library. 3 The collection includes 31 motets (one of which
1 Liber secundus diversarum modulationum singulis, binis, ternis, quaternisque vocibus. Aucto-re Hieronimo Frescobaldo in Vaticana Principis Apostolorum Basilica organista, Romæ, Apud Andre-am Phæum, 1627. See Répertoire International des Sources Musicales, A/I, Einzeldrücke vor 1800, Bänden 1–15, Kassel-Basel-London, Bärenreiter, 1971–2003, Band 3, f 1853.
2 Catalogue of Seventeenth Century Italian Books in the British Library, London, The British Library, 1986, p. 363; oscar Mischiati, Catalogo delle edizioni originali delle opere di Girolamo Frescobaldi, «L’organo», xxI, 1983 pp. 3–82: 65–66; frederick Hammond, Girolamo Frescobal-di, Palermo, L’epos, 2002, pp. 403–405; Saverio franchi, Annali della stampa musicale romana dei secoli xVI–xVIII, vol. I/1, Edizioni di musica pratica dal 1601 al 1650, in collaborazione con orietta Sartori, Roma, IBIMuS, 2006, pp. 559–561.
3 Alexander Hyatt King, Some Recent Acquisitions of Music, «The British Museum Quar-terly», 22, 1960 (1/2), pp. 1–5: 1–2.
Marina Toffetti
6
is divided into Prima pars and Secunda pars): 5 motets for solo voice and basso continuo; 14 for two voices and basso continuo; 6 (one of which is divided into two parts) for three voices and basso continuo, and 6 for four voices and basso continuo. Due to the lack of a part-book, only 14 motets are complete, while 17 of them (one of which is in two parts) are incomplete: all the motets for four voices and basso continuo (6), all of those for three voices and basso continuo (6, one of which is in two parts) and 5 of the 14 motets for two voices and basso continuo. The motet Iesu Rex admirabilis for three voices and basso continuo, incomplete in the Liber secundus, also appears, although with some variants, in the anthology Sacri affetti (Rome, 1625), which is preserved complete. 4 This collection of motets by frescobaldi has so far been published in one modern edition, which includes only the complete motets, and not those missing the part of the cantus secundus. 5 In addition to this a critical edition of the whole Liber secundus, with a reconstruction of the missing part in all incomplete motets, is going to appear as the twenty-sixth volume of the complete works of this composer. 6
The intent of the present article is to show some of the main difficulties encountered during the delicate operation of the reconstruction of the missing part, together with some stylistic characteristics found in frescobaldi’s collec-tion of motets. 7 At the same time, it also aims to stimulate reflection on the
4 Sacri affetti contesti da di veersi [sic] eeclentissimi [sic] autori, raccolti da francesco Sam-maruco, Roma, Luca Antonio Soldi, 1625. RISM B/II 16251. on the differences between the two versions of the motet Iesu Rex admirabilis see Marina Toffetti, Sul processo creativo fresco-baldiano: la duplice versione del mottetto Iesu Rex admirabilis, paper read at the xvI “Colloquio di Musicologia del Saggiatore Musicale” (Bologna, 18th november 2012).
5 Girolamo frescobaldi, Mottetti a 1, 2 e 3 voci con continuo, edizione a cura di Christo-pher Stembridge, Padova, Zanibon, 1987 (Capolavori musicali dei secoli xvII e xvIII); the motet Iesu Rex admirabilis, together with another three motets by frescobaldi included in three different Roman anthologies (Peccavi super numerum, Angelus ad pastores and Ego sum panis vivus), appears in a further modern edition: Girolamo frescobaldi (1584-1643), Mottetti a 2 e 3 voci e basso continuo, a cura di Lorenzo Ghielmi e Mario valsecchi, Bergamo, edizioni Carrara, 1983.
6 Girolamo frescobaldi, Liber secundus diversarum modulationum (Roma, 1627), a cura di Marco Della Sciucca e Marina Toffetti, Milano, Suvini Zerboni, forthcoming (Monu-menti Musicali Italiani, Girolamo frescobaldi, Opera omnia, 26).
7 on stylistic characteristics of frescobaldi’s motets see also Christopher Stembridge, Questioni di stile nei mottetti di Frescobaldi, in Girolamo Frescobaldi nel IV centenario della nas-cita. Atti del convegno internazionale di studi (Ferrara, 9–14 settembre 1983), a cura di Sergio Durante e Dinko fabris, firenze, olschki, 1986 (Quaderni della Rivista Italiana di Musico-logia, 10), pp. 195–213.
Restoring a Masterpiece
7
advisability of carrying out further reconstructions of important scores that are preserved incomplete. 8
Before going into more specific aspects, it is worth devoting a few lines to a preliminary question: why should we devote time and energy to the recon-struction of the missing parts in incomplete polyphony? I can see at least three main reasons why it is important to reconstruct (where possible) the missing parts in incomplete polyphonic compositions:a. to make the compositions which have fallen into oblivion for centuries
become accessible again, both for the purposes of study and research, and for performance and listening;
b. to help refine and modify our traditional view of the history of music, so far mainly (if not exclusively) based on the study of and on listening to complete compositions;
c. to stimulate the analytical investigation of the style of an era, of a genre or of a single composer. 9
Besides that, we must also remember that, contrary to what happens in the restoration of woks of visual art, where the result of a restoration is mostly de-finitive, in the field of music any proposed reconstruction is totally reversible: in case it is considered unacceptable, or even just perfectible, it can be replaced by a stylistically more appropriate reconstruction. In addition to this, the res-toration of a painting, a statue or a building often modifies or even damages the surviving portion of the work of art, while the reconstruction of a missing part of an incomplete composition may not alter or in any way damage the surviving portions of the musical texture. Those who advance doubts regard-ing so-called ‘stylistic’ reconstructions usually maintain that in this field it is very difficult (if not impossible) to obtain satisfactory results. If this opinion
8 In recent times there has been a growing number of editions of music subject to the reconstruction of missing parts: among them see Biagio Marini, Music for Vespers ( february 2004), ed. Thomas D. Dunn (http://www.sscm-wlscm.org/images/WLSCM_Catalogue/ WLSCM_4/Marini_vespers_Critical_Report.pdf); Giulio Cesare Ardemanio, Musica a più voci (Milano, 1628). Composizioni per un’azione pastorale in onore di San Carlo Borromeo, edizione critica a cura di Marina Toffetti, Pisa, eTS, 2012 («Diverse voci…», 11); Carlo Gesualdo, Sa-crae cantiones, reconstruction of the missing parts by James Wood, 2013 (http://www.choroi.net/gesualdo.htm).
9 on the same subject see also Marina Toffetti, The critical edition and the recon-struction of incomplete polyphony, paper read at the international Medieval and Renaissance Music Conference (Certaldo, 4-7 July 2013).
Marina Toffetti
8
can be partly shared, it should also be noted that, in general terms, incomplete compositions are not performed at all, nor studied, with serious consequences for both listeners (who are deprived of listening to a large portion of the Re-naissance and Baroque musical repertoire), and scholars, inevitably brought to base their conception of the history of the music of these eras mainly on the compositions which have survived complete. The reconstruction of incomplete polyphony is undoubtedly only a hypothesis and, as such, it is certainly perfect-ible. 10 Yet an initial hypothesis, if it is deemed inappropriate, will be at least a stimulus to formulate more appropriate ones. The reckoning of advantages and disadvantages of such an operation, even considering all its limitations, is definitely active.
While it is certainly much easier to criticize a hypothesis of reconstruction than to produce a better one, we must admit that the objections to the hypoth-esis of reconstruction can represent a powerful stimulus to the improvement of the quality of reconstructions. By recomposing a missing portion of a com-position, the musicologist often gets too used to his solutions, to the point of becoming accustomed to them, as if they had always existed, while in fact they are only the result of a hypothesis, although reasoned.
The risk that one runs in criticizing a reconstruction is that of referring to single passages, and not to the composition as a whole, forgetting that the lat-ter represents a system within which, as in a chemical formula, it is not possible to change a detail without affecting or, better said, unbalancing many others. Any change often involves many others, so that one of the main difficulties of the work of reconstruction is to compute simultaneously all the consequences of every single change.
In any case, it is much easier to detect that ‘something is wrong’, than to find a satisfactory alternative. However, the mere fact of detecting (even just ‘by ear’) fragments which are stylistically not fully convincing still repre-sents the first step towards a better solution. As long as there are unconvincing elements in a reconstruction, we will not be allowed to stop looking for a better solution.
10 See Biagio Marini, Music for Vespers, Critical report: “The parts cited below as ‘recon-structed by the editor’ are not to be considered definitive but are provided to give a sense of what the pieces may have been like in complete form. Staves and note sizes of these reconstructed parts have been slightly reduced in order to distinguish them from the surviving original parts. Performers are invited to emend or alter the reconstructed parts”.
Restoring a Masterpiece
9
Contrary to what one might imagine, a truly stylistic reconstruction leaves little room for improvisation and oestrus (such weapons can perhaps be suit-able for amateurs, from whom one may expect fanciful reworkings, rather than stylistically convincing reconstructions). A stylistic reconstruction, on the con-trary, is an extremely challenging exercise that requires an impressive knowl-edge of style and compositional techniques, and from which the ‘rebuilder’ himself will never cease to draw benefit. 11 Disassembling and reassembling musical constructs is a puzzle game that involves enormous effort, but effort for which one is invariably repaid. 12
Those who claim that “you will never be able to rebuild the missing part in its original appearance” sometimes do so, albeit unconsciously, to justify any possible solution, even if dictated by free inspiration or fantasy. Since “we can never reconstruct the original version”, and “we will never know if our version corresponds to the original”, then any solution is as good as any other. on the contrary, I am convinced that some solutions are better than others and that, among different hypotheses of reconstruction, some (though perfectible) are acceptable, while others are not.
After considering the (many) advantages and the (very few) disadvantages of the reconstruction of incomplete polyphony, we will now see some of the major problems that each ‘rebuilder’ is inevitably destined to face.
11 See James Wood, Gesualdo: Sacrae Cantiones II. An analysis towards reconstruction, ver-sion 5: July 2011 (http://www.choroi.net/Gesualdo-Introduction.pdf ), p. 6: “And so it was that I had the idea to try and attempt a stylistic reconstruction of the whole set of twenty motets myself. I am quite sure that I had little idea at the outset just how difficult this task would turn out to be, and indeed there were several moments when I was very tempted to concede defeat. However, my determination was fuelled on the one hand by the excitement of bringing these masterful and visionary pieces back to life, and on the other by the stimulation which came from discovering so many secrets within a compositional technique of such phenomenal strength and sophistication, and from which I, as a composer (even 400 years later), could learn so much. And so I persevered with what has turned out to be one of the most challenging and fascinating projects I have ever undertaken”.
12 See Wood, Gesualdo: Sacrae Cantiones II, p. 6: “The scope and extent of this analysis constantly grew and widened during my work on the reconstruction, as I discovered more and more consistent stylistic aspects in the music. These include techniques of counterpoint, text-setting, melodic and harmonic tendencies and rhythmic style. Consequently, the more I worked on the reconstructions the more I learned; and the more I learned, the more I felt the need to review and re-work what I had done previously. This kind of circular process occupied me over a period of more than two years, from february 2008 until March 2010”.
Marina Toffetti
10
1. How to recover the compositional grammar? Considerations on the choice of models
one of the main difficulties to be faced in the reconstruction of missing parts is that of recovering the system of written or unwritten rules of counter-point to refer to. In this regard, we are faced with an apparent paradox. on the one hand, we can not proceed in a stylistic recomposition of a work without defining the frame of the stylistic grammar. on the other hand, such a gram-mar can be inferred only from the compositions themselves, which, in some cases, survive only in an incomplete form. 13
In the case of frescobaldi, for example, it might probably have been easier to rebuild a missing part within a collection of instrumental music, since the surviving complete instrumental collections of frescobaldi are in fact far more numerous. 14 As far as sacred vocal music is concerned, the composer prob-ably published, in addition to the four aforementioned motets included in four anthologies of the time, 15 only two books of motets, one of which (a ‘Liber primus’) has been lost, while the other has survived partially incomplete, so that the complete compositions are too few to constitute a sufficiently repre-sentative sample.
Moreover, it is not at all obvious that the casuistry of compositional behav-iours we can detect in complete compositions (they were also much more nu-merous) can provide all the elements that we need to solve a particular problem of reconstruction. In fact, every compositional problem is closely related to the
13 on the problem of the choice of models see also Wood, Gesualdo: Sacrae Cantiones II, p. 6: “My first decision, then, was what to use as a model. Aware of the significant difference in style between Gesualdo’s madrigals and his sacred music I resolved to refer principally to Sacrae Cantiones I and the Responsoria (as well, of course, as the surviving voices of Sacrae Cantiones II) for my models, and so began a detailed analysis of these works».
14 See oscar Mischiati, Catalogo delle edizioni originali delle opere di Girolamo Fresco-baldi, «L’organo», xxI (1983), pp. 3-82; frederick Hammond-Alexander Silbiger, Fresco-baldi, Girolamo Alessandro, in The New Grove Dictionary of Music and Musicians, Second edition (2004), London, Macmillan, Ix, pp. 238–252.
15 Peccavi super numerum appeared in Selectæ cantiones excellentissimorum auctorum binis, ternis, quaternisq. Vocibus concinendæ, a Fabio Constantino romano insignis Basilicæ S. Mariæ Trans Tyberim musices moderatore, simul collecta, Roma, Bartolomeo Zannetti, 1616, RISM B/I 16161; Angelus ad pastores appeared in Scelta di motetti di diversi eccellentissimi autori, à 2, à 3, à 4 e à 5. Posti in luce da Fabio Constantini romano, Roma, Bartolomeo Zannetti, 1618; RISM B/I 16183; Ego sum panis vivus appeared in Lilia campi binis, ternis, quaternisque vocibus concinnata. A Io. Baptista Robletto excerta, Roma, Giovanni Battista Robletti, 1621; RISM B/I 16213; and Iesu Rex admirabilis in Sacri affetti, Roma, 1625 (cfr. supra, footnote 4).
Restoring a Masterpiece
11
specific context in which it appears, with reference to the texture, to the modal context, but also to the relationship of music with a specific text. Leading these considerations to extreme consequences, one might argue that each composi-tion is a system in itself, and that therefore no surviving complete composition can be of any help in solving the problems posed by incomplete ones. In fact, there are criteria that can apply, mutatis mutandis, also taken out of their spe-cific context, and others for which such an extrapolation would be fatal.
one of the risks that can be run, while rebuilding a musical texture of the past, is that of tending towards a ‘normalization’ of the musical language, of avoiding too imaginative solutions just to follow the ‘rules’ (perhaps it would be better to call them ‘habits’) of the counterpoint of the time in a way too rigorous. By behaving this way one forgets that, in the history of music, inge-nious deviations from the traditional paths have often had more effect than the passive acceptance of a consolidated system of rules.
2. When should the missing part sing and when should it keep silent? Considerations on the texture
The old Italian warning “un bel tacer non fu mai scritto” 16 seems to be par-ticularly suited to those who reconstruct the missing portions of a musical com-position of the past. one of the most insidious questions that may arise is: when should the missing part sing? And when should it remain silent? Although this question may seem trivial at first sight, it is, on the contrary, a very tricky one.
one of the pitfalls that one is likely to face is to feel obliged to fill every gap, as if one were seized with a sort of horror vacui that forces him to produce a score dense beyond measure. of course, in the case of a composition for two voices and basso continuo, the question is usually pretty simple: when the sur-viving voice is silent, it is highly probable that the missing voice sings (since the passages in which the basso continuo proceeds alone, at that time, were generally quite rare and not very extended). In compositions for three voices the problem tends to get more complicated, and in four-voice compositions it may raise questions which are destined to remain unanswered, or at least open to a variety of acceptable solutions. In principle, the abovementioned criterion should serve as a guide: it is better to keep silent than to say some-
16 Literally: “a beautiful silence was never written”. It is a verse from Giacomo Badoaro, Il ritorno di Ulisse in patria,
Marina Toffetti
12
thing unconvincing. If one finds no truly convincing solutions, it is worth con-sidering the hypothesis that in the problematic point the missing voice should be silent. Although it is not advisable to put rests in the missing part for the simple reason that one has not yet found a good solution, the proliferation of unconvincing solutions in a given point is a warning signal that cannot be underestimated.
An analysis of the texture of the whole composition, of the behaviour of surviving voices, of the overall density of the motet (although difficult to assess in the absence of one part) should provide some indications on the compo-sitional style, which can be either more prone to the compactness of Renais-sance polyphony, or to the airy dialogic alternation of the Baroque sacri concerti. A few examples taken from the motets of frescobaldi may be of help in under-standing the problem.
In the motet Corona aurea for four voices and basso continuo, for example, the relatively conventional style and the texture of the surviving parts suggest that the composition, at least in its initial part, could be structured in rather short fragments connected to one another without caesuras and assigned either to the upper semi-chorus (cantus and missing altus part), or to the lower one (tenor and bassus) (example 1, bars 1–3).
The motet Ego clamavi, on the contrary, displays very different characteris-tics. Here the text set to music begins with a heartfelt exclamation, introduced by the personal pronoun ‘ego’. This circumstance, and the melodic morphol-ogy of the opening airy motif entrusted to the cantus primus, suggest the use of a monodic treatment of the same motif (example 2).
During the reconstruction several other solutions were also tested, direct-ed towards a more traditional division of the texture in two alternated semi-choirs. 17 However, repeated attempts to introduce a kind of countersubject to the initial motif could only give rather unsatisfactory results. In the most con-vincing solution, the cantus primus bursts into a powerful and expressive cry (“clamavi”), while the three lower voices are used as a compact entity in a lower register (example 2, bars 1–7).
17 A solution of this type has been adopted by francesco Luisi, who is convinced that “La composizione è improntata all’ordine contrappuntistico e stilisticamente si presenta con procedimenti a coppie di voci contrapposte, riunite in fase di cadenza”. See francesco Luisi, Il Liber secundus diversarum modulationum (1627): proposte di realizzazione della parte mancante, in Girolamo Frescobaldi nel IV centenario della nascita, pp. 163–195: p. 166.
Restoring a Masterpiece
13
A similar exordium, which gives dignity to the solo soprano voice and cre-ates a trade-off between the upper voice and the three lower ones, inaugurates a new way of conceiving the texture that probably recurs also in other places in the same motet. A little further on, at the end of the episode (presumably) in three voices on the words “quoniam exaudisti me Deus”, we can find a further opportunity in which the soprano could sing the words “inclina aurem tuam” as a solo voice, closely followed by the three lower voices engaged to sing the same words in a mainly imitative episode (example 2, bars 7–9).
frescobaldi, however, does not like the schematic reiteration of models, but rather looks for a great variety of solutions. Thus, even in relation to the texture, the pattern of contrast between the upper voice and the three lower voices, which could have been used so far in the motet Ego clamavi, seems to leave room for different solutions in the rest of the composition. The episode entrusted to the three lower voices (example 2, bars 9–11), in fact, could effec-tively be echoed by a brief episode entrusted to the three upper voices (exam-ple 2, bars 12 and following).
furthermore, the soprano voice may not be the only one for which fresco-baldi reserves solo moments and phrases. During the motet, in fact, there are situations in which different voices, in turn, expose patterns with profiles that suggest an accompanied monodic treatment and therefore discourage the in-troduction of ‘countersubjects’ in the missing voice even if it would be possible to elaborate different, rather convincing ones.
Example 1. Corona aurea, bars 1–4
&
V
V
?
?
c
c
c
c
c
1
Co ro na
w Co ro na
w
. œ wau re aœ# œ# wau re a
∑ Co
Co ro na
w
w #su per
∑ su
. œro na au re
œ# œ œ œ# œ œ œ œ œ œ œ œ .œ Jœau re
#
œ œ
ca
wa,
wa,
w
- - - - -
- - - - -
- - - -
- - - - - - -
Marina Toffetti
14
Example 2. Ego clamavi, bars 1–12
&
&
V
?
?
c
c
c
c
c
1 w . œE go cla
w .# œ
# œ œ .œ# Jœ .œ œ# œma vi, cla ma
œ œ# #
w ∑vi,
∑ Œ œquo ni
∑ Œ # œ#quo ni
∑ Œ œquo ni
w w
Œ jœ jœ am ex au dis
# Œ Jœ# Jœ# am ex au dis ti,
Œ Jœ Jœ b œ# œam ex au dis ti me,
b #
- - - - -
- - - - -
- - - - -
- - - - -
&
&
V
?
?
5
œ Jœ Jœ . Jœ Jœ ti, ex au dis ti me, De
Œ Jœ Jœ . Jœ Jœ .œ Jœex au dis ti me, De
∑De us,
. œ
Œ jœ jœ œ jœ jœ us ex au dis ti me, De
œ jœ jœ .b Jœ Jœ .œ Jœus, ex au dis ti me, De
Œ Jœ Jœ œb œ ex au dis ti me, De
b 6
∑ Œ œ in cli
w ∑us,
w ∑us,
w ∑us,
w
œ œ œ œ œ œ# na au rem tu
w# œ œ b
-
- -
- - - - - - - - -
- - - - -- - -
- - -
- --
&
V
?
?
9 w ∑am,
∑ in
œ œin cli na au
∑ Œ œ in cli
w
œ .œ œ œ œ œ œcli na au rem
œ œ œ œ œ œ .œ œ œ rem tu
œ .œ Jœ œ# .œ œ œ œ œna au
œ .œ Jœ œ# .œ œ œ œ œ
.œ jœ w #tu am,
. œ am,
œ œ wrem tu am,
œ œ w
Œ œ œ œet ex au di, et
Œ # œ # œ œet ex au di, et
Œ œ et ex au di,
. œ # 6
- -
-
- - -
-
-
-
- - - -
- - - -
- -
- -
- -
-
Restoring a Masterpiece
15
In the collection there is another motet, the text of which begins with the personal pronoun ‘ego’: Ego sum qui sum. Also in this motet it is likely that, at least at the very beginning of the composition, a single voice began to sing (example 3, bars 1–4).
Contrary to the motet Ego clamavi, which, as seen, appears to have been conceived as a concertato motet, with frequent solo episodes and a great vari-ety of contrapuntal and scoring solutions, the motet Ego sum qui sum appar-ently presents a more schematic course, with the frequent alternation of pairs of voices that proceed mostly parallel at the distance of a third, a tenth or a sixth. See, for example, the episode built in correspondence with the words “et consilium meum”, entrusted to the soprano-tenor couple (which moves at the distance of one tenth), probably followed by the alto-bass couple in correspon-dence with the words “non est cum impiis” (example 3, bars 4–8).
Example 3. Ego sum qui sum, bars 1–8
&
&
V
?
?
c
c
c
c
c
1. œ w
E go sum
. œ E go sum
. œ E go sum
W
wqui sum,
w #qui
b wqui sum,
b w
. œ e go sum
. œ sum, e go sum
. œ e go sum
. œ E go sum
w w
# qui sum et
wqui sum
#qui sum et
wqui sum
w#
-
- -
-
-
-
&
&
V
?
?
5 œ œ
con si li um
# œ œcon si li um
w# . œ
.œ œ .œ œ .œ œ .œ œ œ œ œ œ me
.œ œ .œ œ .œ œ .œ œ œ œ œ œ me
w ∑um
w ∑um
non est cum
w
.im
w
- - -
- - -
- - - -
- - - -
Marina Toffetti
16
3. Where does the obligation finish and where does the fantasy begin? Considerations on style
In a few contrapuntal situations it is possible to find only one really sat-isfactory solution. These situations, to be honest, are not so many. In general, they can be found in a context of strict (or, at least, presumably strict) imita-tion, mainly (but not exclusively) at the beginning of the motet, when all sur-viving voices exhibit the same motif at well established intervals of time and pitch. I cite, for example, the opening section of the motet Vox dilecti mei for two sopranos and continuo, in which the first cantus initially exposes the main motif, presumably expecting the second to sing a ‘real answer’ (example 4).
Example 4. Vox dilecti mei, bars 1–4
even in the context of imitation, however, there are several circumstances in which a seemingly obvious solution can be complemented by others, sig-nificantly different, which deserve to be scrutinized with the utmost care. This happens, for example, in the first imitative episode of the motet Tota pulchra es. Here, the first motif should be entrusted to the now lost cantus secundus.
At first glance, there would be no doubt in identifying in the motif D-D-f-G-D the more suitable antecedent to the surviving response of the tenor voice (A-A-C-D-A). However, we must consider that here frescobaldi is using the initial motif of the Marian antiphon (D-D-f-G-A), also used in the homonymous motet by Palestrina included in his famous fourth book of motets for five voices. 18 If we assume this, the response of the tenor would then
18 on Palestrina’s motets on the Canticle see Jessie Anne owens, Palestrina as a reader. Motets from the “Song of Songs”, in Palestrina e l ’Europa. Atti del III Convegno Internazionale di Studi (Palestrina, 6–9 ottobre 1994), a cura di Giancarlo Rostirolla, Stefania Soldati, elena Zomparelli, Palestrina, fondazione Giovanni Pierluigi da Palestrina, 2006, pp. 263–287.
&
&
V
?
b
b
b
b
c
c
c
c
1 .w me di
.w b
# lec ti me i
∑ wVox
∑ Vox
w w4 3
Œ œ œ œ œ œ œ œ œpul san
ndi lec ti
b . œdi lec ti
b w6 7 6
tis;
œme i
wme
- - -
- - - - - - -
- -
Restoring a Masterpiece
17
be a ‘mutation’ of the initial motif, justified by the need to prevent the conse-quent from going outside the proper modal context, according to a widespread process which had also been used by Palestrina in the same motet. Let us look at the two possible solutions 19:
Example 5a-b. Tota pulchra es, bars 1–4
A different case is that of the imitative opening section of the motet Vidi speciosam for two sopranos and basso continuo. Here, after some rests, the can-tus primus sings the words “vidi speciosam sicut columbam” (example 6a) There is no doubt that the cantus secundus should come first and sing the same motif. But things are not so simple.
If we just ‘cut’ the motif exposed by the cantus primus and ‘glue’ it in com-bination with the basso continuo at the beginning of the motet, in correspon-dence with the words “sicut columbam” it will produce a completely different harmonic situation, which is very far from the harmonic habits of the early seventeenth century (example 6b). In other words, the literal recovery of the whole motif generates problematic vertical situations. If ever we wanted to keep the characteristic interval of the semitone at the word “columbam”, we
19 Both solutions are compatible with the basso continuo. The absence of the cifra 6 above the f of bar 2 should not be regarded as binding in the reconstruction, as it would be vice versa its presence, since the numerica of the figured bass is very often incomplete.
&
V
?
c
c
c
1 w
To ta
w b
wpul chra es,
∑ To
. œ
ta pul chra
w b
wes,
.w
- -
- -
&
V
?
c
c
c
1 w
To ta
w b
wpul chra es,
∑ To
. œ
ta pul chra
w b
wes,
.w
- -
- -
a
b
Marina Toffetti
18
should be forced to further subdivide the motif “sicut columbam” into two smaller fragments (“sicut” and “columbam”). However, in frescobaldi’s motets known to us, the composer never shows so minutely analytical an attitude (example 6c).
As the verse “vidi speciosam sicut columbam” is composed of two smaller entities, “vidi speciosam” and “sicut columbam”, a possible solution, accord-ing to the line of the continuo, could be that of taking literally the individual motifs of the cantus primus (“vidi speciosam” and “sicut columbam”), but changing the interval between them (example 6d). As a further alternative, it could also be possible to postpone the fragment «sicut columbam» in the cantus secundus.
&
&
?
c
c
c
1
w Vi di
W
. œ œ œ œ œspe ci o
. œ 6
w Vi di
œ œ œ œ sam sic ut co lum
w 4 3
. œ œ œ œ œspe ci o
œ œ œ œ œ œ œbam, sic ut co lum
. œ œ œ6 5
œsam
œ œ# bam
œ4 3
- - - -
- - - - - - - - - -
&
?
c
c
1
W
. œ 6
w Vi di
w 4 3
. œ œ œ œ œspe ci o
. œ œ œ6 5
œ œ œ œ sam sic ut co lum
œ œ# 4 3
- - - - - - -
&
?
6
œ œ œ œ œ œbam, sic ut co lum
6 5
œ œ# bam
4 #3
- - - -
&
&
?
c
c
c
1
w Vi di
W
. œ œ œ œ œspe ci o
. œ 6
w Vi di
œ œ œ œ sam sic ut co lum
w 4 3
. œ œ œ œ œspe ci o
œ œ œ œ œ œ œbam, sic ut co lum
. œ œ œ6 5
œsam
œ œ# bam
œ4 3
- - - -
- - - - - - - - - -
a
b
c
Restoring a Masterpiece
19
Example 6a-b-c-d. Vidi speciosam, bars 1–7
So far we have considered the first part of the problem that we set (where does the obligation end?). However, the most difficult aspect of the reconstruc-tion is what to do at those points – and there are many – at which the composi-tion is no longer subject to a strict system of rules, but requires ingenious ideas or original and sometimes bizarre solutions. The reconstruction of the most free points of the composition should be based on a deeper knowledge of the author’s style, in the awareness that, in these circumstances, the probability of identifying the solution adopted by the author decreases.
4. Does the missing part repeat a motif which has already been sung by another voice, or does it introduce a new motif? Considerations on the contrapuntal structure
Another pitfall inherent in the reconstruction is to consider the missing part as if it were in some way gregarious compared to the surviving ones, with-out considering that, in the original composition, it had exactly the same dig-nity as all the other voices. When you recompose the missing part, however, it happens that you unconsciously let the surviving parts lead the game by introducing into the missing voice repetitions of motifs that have already been sung by the other voices, forgetting that the missing part could also anticipate what the other voices would then repeat. This happens, for example, in the motet Vox dilecti mei, in correspondence with the words “aperi mihi”, which probably have to be tuned for the first time by the cantus secundus (example 7).
Moreover, we must also keep in mind that what survives in the remaining parts could represent the varied repetition of a motif which could have already been tuned for the first time, in a slightly different way, by the missing part.
&
&
?
c
c
c
1
w Vi di
W
. œ œ œ œ œspe ci o
. œ 6
w Vi di
œ œ œ œ sam sic ut co lum
w 4 3
. œ œ œ œ œspe ci o
œ œ œ œ œ œ œbam, sic ut co lum
. œ œ œ6 5
œsam
œ œ# bam
œ4 3
- - - -
- - - - - - - - - -
d
Marina Toffetti
20
So when the missing part tunes a motif that will be repeated by the other voices, this motif may be slightly different. In the light of these observations, during the reconstruction it is advisable to alternate a linear process (from the beginning to the end of the composition) with frequent leaps forward and flashbacks.
5. Which text should the missing part sing? Considerations on the text-music relationship
We know that the point of departure of the composition of a motet was the reading of a text, from which sprouted the composer’s motivic ideas. Similarly, the text should guide us in every phase of the reconstruction. In the aspects that effect the overall structure of the composition, the main sections of the text will help us to identify the main cadences and the more prominent musical caesuras. In the minute details of the composition, the correct accentuation of the words and their effective declamation will guide us in the reconstruction of important rhythmic and melodic aspects. In his collection of motets, fresco-baldi always pays very special attention to the correct accentuation of words and to the effective declamation of the text, so that there is a particular focus on the role of rhythmic and metrical position (also in relation to the tactus) of each single word. This feature of frescobaldi’s writing imposes the maximum rigour also in the recomposition of the missing part.
Example 7. Vox dilecti mei, bars 6–8
&
V
V
?
b
b
b
b
6 A Œ œa tis; a
Œ .œ Jœn œ a pe ri mi hi,
.œb œ œ w tis;
wb 4 3
Jœ Jœ# œ pe ri mi hi,
∑ Œ œa
∑ Œ .œ Jœ# œ#a pe ri
7 6 6 7 6#
œ œ œ œ œ œ œ œ œ œ œso ror me a,
Jœ Jœ# œ pe ri mi hi,
œ œmi hi, so ror
6
- - - - - -
- - - - - -
- - - - -
Restoring a Masterpiece
21
6. Simple or double counterpoint?
A further issue of some importance, when rebuilding the missing part, concerns the choice of the contrapuntal style to be adopted in different situ-ations. In particular, there are a few situations in which it seems possible to use rather elaborate techniques of double counterpoint, due to the fact that some motifs presented in succession prove to be superimposable and usable
Example 8a-b. O bone Iesu, bars 17–24
b
a
&
V
?
b
b
b
17
w ∑am
w nam vul tus
w n
w vul tus
# wtu
w# 6 5
w# Œ jœ jœtu i re pro
#i et no li
w #
Jœ Jœ Jœ Jœ œb œ œ jœ jœba re a di lec tis tu is, a di
œ œ œ Jœ Jœme re pro ba re a di
. œ b5 6
- - - - - - - - -
- - - - - - - - -
&
V
?
b
b
b
21
œ œ .œ jœ lec tis tu is, et
.œ jœ wlec tis tu is,
œ œ
# œ œno li me re pro
Œ jœ jœ Jœ Jœ Jœ Jœ œb œre pro ba re a di lec tis
œb œ#
œ Jœb Jœ œ œ ba re a di lec tis tu
œ Jœ Jœ œ œ .œ jœtu is, a di lec tis tu
œ œ
wis
is,
œ
- - - - - - - - - - -
- - - - - - - - - - - -
&
V
?
b
b
b
17
w ∑am
w nam vul tus
w n
w vul tus
# wtu
w# 6 5
w# tu i et
#i et no li
w #
œ œ œ œ no li me re
œ œ œ Jœ Jœme re pro ba re a di
. œ b5 6
-
- - - - - - - - -
- - -
&
V
?
b
b
b
21
œ œ Œ jœ jœpro ba re, re pro
.œ jœ wlec tis tu is,
œ œ
Jœ Jœ Jœ Jœ œ# œ ba re a di lec tis tu
Œ jœ jœ Jœ Jœ Jœ Jœ œb œre pro ba re a di lec tis
œb œ#
œ œ# œ jœb jœ œb œ is, a di lec tis tu
œ Jœ Jœ œ œ .œ jœtu is, a di lec tis tu
œ œ
wis,
is,
œ
- - - - - - - - - - - -
- - - - - - - - -- - - - -
Marina Toffetti
22
such as a subject and as a countersubject. In some cases, the lack of the cantus secundus (or of the altus) complicates matters, especially when the same point can be solved in double counterpoint, but also using much simpler imitative techniques without using any superposition of motifs. If we look, for example, at the episode of the motet O bone Iesu corresponding to the words “et noli me reprobare a dilectis tuis”, we will notice that, using a relatively simple imitative technique, it is possible to reconstruct the missing part in a rather convincing way (example 8a). The same passage, however, is compatible with a much more elaborate solution, in which the motif used on the words “et noli me”, which presents rather long values of duration, convincingly overlaps with a more live-ly motif, employed by frescobaldi in correspondence with the words “reprobare a dilectis tuis” (example 8b).
Another example of this problem could be found in the initial episode of the motet Ego flos campi, in which a motif used as a subject could be accompa-nied by another motif employed as a countersubject.
of course, in the presence of different solutions the evaluation of which is the most convincing should be conducted case by case. However, it should be noted that, in principle, it seems implausible that frescobaldi could have conceived such complex constructs, and then changed his mind and adopted much easier solutions.
Conclusions
The arduous task of reconstructing the missing part in incomplete po-lyphony proved to be as exciting as valuable an experience. As we have seen, falling within the philological operations belonging to the field of emendatio, the reconstruction of an incomplete score is just a working hypothesis, can not damage the surviving portions of the composition and can at any time be replaced by a better reconstruction.
Although it is appropriate that, in the event of a publication, a single musicologist takes responsibility for the result of the reconstruction of the en-tire missing part, several recent experiences have shown that the reconstruc-tion can be extremely well facilitated by a collective approach, in which the members of a working group proceed simultaneously, with frequent exchanges
Restoring a Masterpiece
23
of problems and solutions, engaging in a kind of collective brain-storming. 20 It is useful to exchange suggestions in real time, submit specific problems to col-leagues, compare different solutions, detecting, for each hypothesis, strengths, weaknesses, acceptable and unacceptable aspects and mitigate the inevitable excitement that occurs when you think you have found ‘the’ solution to a prob-lem by comparing it with other solutions displaying different characteristics.
All this, ultimately, can help to stimulate an increasingly qualified debate on the minute details of an individual style observed from the very inside of every composition. If properly set up, such a work can contribute to the draft-ing of a relevant and meaningful grammar of style (a ‘thesaurus’), enabling a community of experts to evaluate which solution is better and which worse, what is acceptable and what is inadequate, and contributing significantly to our knowledge of the style of an epoch, of a genre or of a single composer.
Translated by Paweł Wróbel
20 Collective experiences of reconstruction have already been experimented with in some international research projects focused on the recomposition of incomplete polyphony. In this regard, see: Atelier Virtuel de Restitution Polyphonique of the Centre d’Études Supérieures de la Renaissance and of the université françois Rabelais in Tours (http://ricercar.cesr.univ-tours.fr/3-programmes/eMN/Ateliervirtuel/); and Digital Du Chemin (http://duchemin.haverford.edu).
Marina Toffetti
Abstract
The intent of the article is to show some of the main difficulties encountered during the delicate operation of the reconstruction of the missing part of the cantus secundus in the Liber secundus diversarum modulationum by Girolamo frescobaldi, together with some stylistic characteristics found in this collection of motets. At the same time, it also aims to stimulate reflection on the advisability of carrying out further reconstruc-tions of important scores that are preserved incomplete.The reconstruction of the missing parts in incomplete polyphony makes the works accessible again—both for the purposes of study and research, and for performance and listening; refines and modifies our traditional view of the history of music; and stimulates the analytical investigation of the style of an era, of a genre or of a single composer.The author discusses several analytic, stylistic and formal questions arising during reconstruction of the missing parts in incomplete polyphony: How to recover the compositional grammar? When should the missing part sing and when should it keep silent? Should the reconstructor follow the contrapunctual rules, or is there a place for more imaginative solutions? Does the missing part repeat a motif which has already been sung by another voice, or does it introduce a new motif ? Which text should the missing part sing? Simple or double counterpoint is more appropriate?
Keywords: Girolamo frescobaldi, baroque music editing, reconstructing incomplete polyphony.
